A data fitting approach to series convergence acceleration

The paper deals with the transformation of real number series \(\sum a(n)\) (1\(\leq n\leq \infty)\) by use of approximations of the form \(a(n)=\sum b(k)f(k;n)\) (1\(\leq k\leq r)\) where the series \(\sum f(k;n)\) (1\(\leq n\leq \infty)\) have, for \(1\leq k\leq r\), known sums and the coefficients b(k) are obtained by least squares fitting. Numerical examples concerning the transformation of series such as \(\sum (cn+d)^{-2}\) (1\(\leq n\leq \infty)\), f(k;n) having the form \(n^{-k-1}\), and others in which f(k;n) has the form \(\exp (-n)L_ k(n)\) (L denoting a Laguerre polynomial) are provided.

Recommendations

Cites work

Cited in

(1)

Irregular input data in convergence acceleration and summation processes: General considerations and some special Gaussian hypergeometric series as model problems

This page was built for publication: A data fitting approach to series convergence acceleration

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q805129)